Make a Real-World Impact Through Data Visualization by Understanding its History and Recent Trends
Authored by Andy Krackov
The idea of visualizing data — that is, of transforming numbers into illustrations to make findings accessible to all — had to start somewhere, right? Turns out we actually can point to a specific time when data visualization, best as we know, had its beginning.
This is the first published bar graph, created by William Playfair in 1786 for an atlas on Scottish trade.
Playfair gets credit, too, for coming up with the idea for the line/trend graph and the pie graph. Not too shabby for a career!
The map below, too, played a key role in data visualization. In Soho London in 1854, there was a severe cholera outbreak, which we perhaps wouldn’t have known about, except that John Snow determined that it was foul water — not bad air, as some presumed — that caused cholera. The map that he created, where each house represents an individual hit by cholera, quite literally changed how we “saw” disease (the pump on Broad St. that he refers to was the source of the cholera outbreak).
By the 1950s, data visualization was common enough in American newspapers that at least one — the New York Times — felt comfortable enough with data viz as a common vernacular to put together a compilation of graphs that we today would refer to as a dashboard.
(You can see more of these in Stuart Thompson’s blog post about New York Times graphics from the 1950s).
These are early examples of the important role data visualization played in shaping social change. I’d argue, however, that data visualization hasn’t transformed greatly since these milestones. Yes, nowadays we’re regularly churning out beautiful data interactives that would make William Playfair’s head spin. But on a more fundamental level, is data visualization being used in the social sector as a tool to educate and persuade? Are the maps and graphs we create truly raising awareness of issues, informing policy decisions, and empowering communities?
On that score, I think the record is a bit more spotty. In essence, we’ve learned over the years that communicating with data is no easy task; you can’t just create a simple graph and put it on auto-pilot to do your bidding. There are, however, some important trends taking place these days that we can point to and collectively learn from. Here are some key developments that I’m observing with data visualization these days:
Turning Data into Games and Intellectual Exercises
Who says data always has to be so serious? I’ve been noticing a trend where organizations transform numbers into games (or intellectual exercises if you want to be a bit more refined in describing it). Take this New York Times interactive where they invite you, the reader, to guess at trends in cause of death, first for HIV, guns, and car accidents and then drug overdoses, then seeing how close you are to guessing the movement of the line graph.
When you think about it, the concept is really basic: It’s just a trend graph that you’re drawing, after all. But the point is that you’re drawing it, not just observing it, and in so doing, may be more likely to retain the information. A simply executed data game is this 538 quiz on public opinion regarding guns; another is one created by Velir working with one of our clients — the Commonwealth Fund — that asked, and had the user answer the question of what would it mean for individuals if health care in your state improved. All of these point to this trend of jumping a bit out of our comfort zone of data always being so serious, and transforming numbers into an opportunity for some play.
Making Large Numbers Seem Relatable
Turns out us humans are not very good at truly understanding the size of a large number like one million, or even ten thousand. You might think we would understand such a scope from reading these numbers so frequently in the media. Trouble is, we don’t typically “see” such numbers in our day-to-day lives. As a result, big numbers can seem like abstractions to us, where we understand them intellectually, but don’t really “feel” them. Given this, there’s a growing trend in data visualization to help readers better relate to large numbers. Take this example from the Washington Post, where you can click on figures as you scroll down the page of people to see what 1,077 individuals killed by mass shootings really looks like.
On a more playful topic, The Pudding looked at the (unlikely) odds of making it big in the music industry in New York City, where roughly 3,000 bands played multiple shows over a three-year period. And the Wall Street Journal showed the reader the tiny chances of winning the lottery by letting him/her scroll through 292 million dots on a screen. As we build out our data visualizations, it’s important to remember that simply saying a large number may not have the impact you want. It may be essential, too, to “show” that large number.
Animating Data
We’ve no doubt all come across instances in which data “move” in some way or another — a trend graph that builds from the starting year, bar graphs that appear one at a time. I used to think the concept of animating data was a gimmick, but then I saw the impact that it could have. When I was working at the Lucile Packard Foundation for Children’s Health, for example, I used to show a presentation of the decline in teen births in counties across California. Google made it easy for me to animate a bar graph to show the change over time (in essence, you’re looking at a bar graph getting bigger or smaller), and I could tell in my presentations how that simple “trick” of animation was helping people better understand, and retain, the data.
Similarly, when I was at the California Health Care Foundation, my communication colleagues began putting together an annual visualization summarizing historical trends in national health care spending that became familiar — even anticipated — by many who work in health care policy. As time marches on in this visualization, trends become clear, perhaps clearer than if these data are presented as a simple line graph. Pay attention, for example, to the orange-shaded area, which represents out-of-pocket spending. You can see how significantly that declines from 1961 to 2016.
Speaking of health care, here’s another example of data animation, where the animations aren’t as central to the story, but as you scroll through this essay on why the US leads the world in health care spending, you see how thoughtfully designed smooth animations of visualizations help tell this story in a more effective way that keeps the reader engaged.
Humanizing Numbers
Another trend to keep tabs on is humanizing data. Historically, we’ve tended to bifurcate data and stories, as if you can have either one of the other, not both. But that’s not the way we process information. In some cases, we’re persuaded by facts, and in other instances, the personal — e.g. a poignant photo, a quote that helps color the data, a video of an affected family — helps make the case. Given that, it’s important that we paint data on a broader canvas with story elements that can help humanize the numbers. There’s no magic formula here. Some data stories work well with facts being the primary storytelling lens, while with others, stories carry the day, with data inserted here and there to help add clarity.
My organization, Velir, for example, worked with a communication firm, Purpose, to demonstrate how many US employees are shift workers — meaning they don’t know their schedule from week to week. We primarily told this story through numbers, but Purpose had quotes and even videos that we could insert into the package to help demonstrate the issues that shift workers face.
Reader Input
I note above the trend of turning data into games, in order to make for a more participatory experience. There are simpler ways, too, to invite the reader to transform the data content. In this Vox article, for example, the reader selects the year he/she was born in order to change the data in the story.
The Washington Post takes a similar approach with this story on the gender pay gap. Scroll down to choose a profession and see related data. By offering a curated — yet tailored — experience, these stories make it more likely that the reader will act on the information, given that they’ve been invited to personalize it to their own interests.
You may have remarked to yourself that many of these examples were created by major news outlets, which have expertise in storytelling and, increasingly, a stable of interactive artists and data visualization specialists to whom they can turn. There’s no question that news outlets like the Washington Post and the New York Times, among others, are leading the way with visualization, doing things with data that other social sector organizations would find difficult to accomplish. But we shouldn’t necessarily try to mimic — and when we can’t, just give up in being creative with data display. We’d do better instead to learn from the leaders and transpose these trends, even in rudimentary ways, to our own work. We’re always eager to help organizations puzzle through what they can do, so you’re welcome to reach out to us.
Finally, there’s one important trend that we all can implement: Keep it practical. Talk to your users, for example, before you build to understand how they want to engage with the information. In Velir’s data work, we’re hearing more and more from clients who want to create printable views of data — e.g. a one-page PDF fact sheet with contextual information and charts. The thinking here is that, while beautiful interactives look amazing and play an important role, our clients also need something tangible to hand out to people during meetings and take back to their office. Even in this digital age, there still are very few things that can beat interacting with a piece of paper. This may be among the more interesting developments — that is, the more we push the digital envelope, the more we may recognize that the tried-and-true still has an important place.
ABOUT THE AUTHOR
Andy Krackov is the VP for Data Strategy at Velir. He has developed a specialty as a data content strategist who can translate data to multiple audiences, in order to inform policymaking, raise public awareness of issues, empower communities, and achieve other meaningful social change. He has served on national and state boards related to data, speaks at leading conferences about effective use of data, and has written for multiple publications on the topic. He began his work as a journalist at U.S. News & World Report, and has a bachelor’s degree in history from Washington University, as well as a master’s degree in communication theory from Stanford University.
ABOUT VELIR
We are a fully-integrated digital agency based right outside of Boston, in Somerville MA. With a team of over 130 digital professionals, we are best-known for our digital work for some of the world’s largest and most influential non-profit organizations, membership associations, and Fortune 500 enterprises. Our capabilities in marketing technology and data visualization are recognized as the best in the industry, stemming from a 17-year focus on digital strategy, creative design, and web engineering and application development. We have completed complex implementations and large-scale builds for brands including Bayer, the Robert Wood Johnson Foundation, the Metropolitan Museum of Art, AARP, Yale, and Informa. We have been honored with the recognition of “Small Business of the Year” by both the Greater Boston Chamber of Commerce and the Somerville Chamber of Commerce, and have offices in Boston and Washington DC. For more information, please visit www.velir.com.
